ABC | Volume 110, Nº1, January 2018

Original Article Santos et al Development and psychometric validation of HIPER-Q Arq Bras Cardiol. 2018; 110(1):60-67 Of these patients, 101 (54.9%) were retired. Participants’ characteristics are described in Table 1. Development of HIPER-Q The literature review on health education for hypertensive patients in CR programs revealed consistent findings between the articles. The first version of the HIPER-Q was developed based on literature data. Nineteen items were constructed encompassing seven important domains in patient education: self-care, treatment, diagnosis, physical exercise, concept and pathophysiology, signs and symptoms and risk factors. Similar to other educational instruments, 12,31 for each item, one answer is considered the “most correct” one and receives score 3, and another answer is considered “partially corrected” and receives score 1. The other two answer options – the incorrect option and the “don’t know” option receives no score (zero). According to the classification described in Table 2, the sum of the scores represents mean total knowledge, where the maximum score of 51 points corresponds to ‘perfect’ knowledge. Clarity validation The construction rules of the item sources and of the theoretical analysis of the items, content and semantics were considered ‘clear’ by 79% of the specialists, with a median clarity score of 8.5 (0.75). However, most of the items received comments on their semantic contexts. Each item was widely discussed by the authors, and all changes suggested by the specialists were accepted. This version of the questionnaire was analyzed by the same professionals, and the final version was then provided, with 96% of agreement between the items and median clarity score of 9.54 (0.30). Pilot study The average time for completion of the questionnaire by the participants (n = 30) was 15.4 ± 2.2 minutes. The median clarity score was 8.7 (0.25), and no item had a clarity score lower than 7.0, indicating that the questionnaire was well understood by the target population. Test-retest reproducibility Total ICC of the instrument was 0.804, obtained by the final test-retest scores. 27 The items “Also with respect to systemic arterial hypertension, we can affirm that” and “What is the best diet for patients with systemic arterial hypertension?” had a ICC lower than 0.7 (0.43 and 0.58, respectively) and were excluded from the final version, 27 which was then composed of 17 questions. The ICC of each question is presented in Table 3. Psychometric validation The HIPER-Q was administered to participants of CR programs, and the mean scores of the questionnaire items are shown in Table 3. Overall, the HIPER-Q showed a moderate internal consistency (Cronbach's alpha = 0.648). With respect to criterion validity, a relationship of HIPER-Q total score was found with educational attainment and family income. Weak positive correlations were found of knowledge level with educational attainment (rho = 0.346; p < 0.01) and family income (rho = 0.176; p = 0.017). Dimensional structure was evaluated by exploratory factor analysis. The Kaiser-Meyer-Olkin (KMO = 0.669) test and the Bartlett’s sphericity test ( X 2 2066.56; p < 0.001) indicated adequacy of data for factor analysis. Five factors were extracted and, together, they accounted for 51.1% of the total variance of the items, whose characteristic values were > 1.1. Table 4 displays the factor loadings of the items. Factor “1” reflects “General Conditions”, and is responsible for 18.8% of total variance, whereas the other factors had a lower influence of the variance. Factor “2” reflects “Treatment”; factor “4” reflects “Physical Exercise”; factor 4 reflects “risk factors” and factor 5 reflects “self-care’. Descriptive analysis The instrument had a median total score of 26 (10). In patients’ classification, a high prevalence (44.6%) of “acceptable knowledge” was observed. Patients showed greater knowledge about the items: “If a health professional says that your blood pressure is altered, you should”, “On the basis of your knowledge about systemic arterial hypertension, answer the following:” and “Which of the risk factor groups below has the greatest influence on the development of systemic arterial hypertension?”. The lowest level of knowledge was seen for the items: “With respect to self-measurement of blood pressure, it is correct to say that”, “About the white coat syndrome , it is correct to say that” and “Which among the items listed below are the most accurate in the diagnosis of systemic arterial hypertension?”. Regarding the knowledge domains, patients showed higher level of knowledge in the areas – “disease” and “concept and pathophysiology”. On the other hand, the lowest level of knowledge was shown for the “diagnostic” and “signs and symptoms” domains. As shown in Table 1, greater knowledge about SAH was associated with coronary artery disease (p < 0.001), dyslipidemias (p = 0.006), myocardial infarction (p < 0.001) and peripheral obstructive arterial disease (p = 0.004). In addition, previous angioplasty (p < 0.001) or cardiac surgery (p = 0.002) was associated with greater knowledge about the disease. Discussion Patient’s education is one of the central components of CR, and is crucial for promoting the understanding about secondary prevention strategies and adherence to treatment. 9,28,31 In the present study, a new tool for the assessment of knowledge in hypertensive patients enrolled in CR programs was developed and psychometrically validated by a rigorous process. In general, clarity, internal consistency, reliability, dimensional structure and criterion validity were established, indicating the validity and usefulness of the HIPER-Q in the assessment of hypertensive patients’ knowledge about the disease. The first data to be considered is the clarity index, generated by professionals and patients, demonstrating that the instrument proposed can be easily understood 62